EleutherAI ML Perf Reading Group: DeepSeek V3

For session 7 of the Eleuther AI ML Scalability & Performance reading group, I presented the DeepSeek V3 paper, and also covered parts of DeepSeek V2 for comparison.

My annotated versions of these papers can be found be found on my Github here.

Papers:

1. DeepSeek V3
2. DeepSeek V2

Note: you may have to disable ad blocker for the YouTube player to render correctly. Alternatively, you can watch the recording directly on YouTube here.